Project

AIGuard is an open-source, MIT-licensed toolkit maintained by a community of contributors. It is local-first by design: no SaaS account, no telemetry, no required external services.

Links

• Roadmap — where the project is going next
• Releases — released versions
• Contributing — how to get involved
• GitHub repository
• PyPI package

Philosophy

Test before deployment. Monitor after deployment. AIGuard treats LLM safety as engineering — measurable, reproducible, and testable in CI.